Energy metabolism in the perfused, arrested rabbit heart

J Mol Cell

Cardiol21,211-221

Energy

(1989)

Metabolism

in the Perfused,

G. Kotsanas,

C. L. Gibbs

Arrested

Rabbit

Heart

and I. R. Wendt

Department of Physiology, Monash University, Clayton, 3168, Australia (Received 11 May 1988, accepted in revisedform 12 October 1988) G. KOTSANAS, C. L. GIBBS AND I. R. WENDT. Energy Metabolism in the Perfused, Arrested Rabbit Heart. Journal of Molecular and Cellular Cardiology (1989) 21, 211-22 1. Energy metabolism of quiescent cardiac muscle was studied in the isolated rabbit heart preparation perfused at constant pressure by the Langendorfftechnique. Oxygen consumption (Mvo*), coronary flow rate (CFR) and the steady state concentrations of high energy phosphate compounds were determined in hearts reydered asystolic using modified Krebs-Henseleit (KH) media containing 11 rnM glucose as substrate. Basal MVo2 and CFR were significantly higher in hearts arrested by Ca ‘+ depletion (low Ca KH) compared to Kf excess (high K KH). Substitution of glucose in low Ca KH with a mixture containing glutamate, fumarate and pyruvate (low Ca KH + GFP) resulted in a 25% increase in the basal Mvo2 but a 20% decline in CFR. Supplementing the low Ca perfusate with 30 g/l dextran (low Ca KH + dextran) depressed both the basal Mpo2 (350/ o ) and CFR (75%). Differences in the basal MPO, under the different perfusion conditions were not accompanied by significant changes in the tissue levels of ATP, CrP or Cr. Compared to low Ca KH arrested hearts, those perfused with low Ca KH + GFP or low Ca KH + dextran did, however, show significantly lower tissue levels of ADP, AMP and Pi, but higher cytosolic ratios of [ATP]/[ADP][P,] and [CrP]/[Cr][Pi]. A s a consequence of the higher phosphorylation potential the free energy of ATP hydrolyis increased. There was no significant difference in any of these parameters between high K KH and low Ca KH perfused hearts. It is concluded that in the perfused, arrested heart none of the parameters that are used to describe the myocardial energetic state, e.g. free [ADP] or the cytosolic

[ATP]/[ADP][P,] ments. KEY energy

ratio, uniquely correlates with the basal metabolic rate as estimated from MVO,

WORDS: Basal metabolism; phosphates; Phosphorylation

Perfused heart; Ca2+ depletion; K+

An unusual feature of quiescent cardiac muscle is its high metabolic rate. It is high relative to that of other muscle types and also constitutes a large fraction of the total (basal + active) metabolism of the working heart (Gibbs, 1978; Chapman et al., 1982). The nature of the metabolic processes that account for myocardial basal metabolism, as well as their fractional contribution to the basal metabolic rate, remain poorly understood (see Gibbs, 1978; Gibbs and Chapman, 1979; Loiselle, 1987, for reviews). In addition, it is difficult to define the basal metabolic rate unambiguously because of the number of external variables that influence its magnitude (Lochner et al., 1968 ; Gibbs and Kotsanas, 1986). For example, the resting heat rate of isolated papillary muscles exhibits a pronounced time-dependent decay in magnitude following isolation of the muscle (Chapman and Gibbs, + 11 $03.00/O

Oxygen

consumption;

High

potential.

Introduction

OOZZ-2828/89/020211

excess;

measure-

1974; Loiselle and Gibbs, 1979) and also varies with metabolic substrate (Chapman and Gibbs, 1974). In whole heart studies the oxygen consumption (Mpo,) of the asystolic perfused heart has been shown to depend in part on the method of cardioplegia (Penpargkul and Scheuer, 1969) as well as on the perfusion pressure and flow rate in the coronary vessels (Arnold et al., 1968; Lochnelr et al., 1968). A recent report from this laboratory which examined in detail the influence of perfusion conditions on the basal Mvo, of the Langendorff perfused rabbit heart confirmed the substrate dependency first observed in papillary muscle studies and demonstrated the influence of coronary flow rate (CFR) on the basal metabolism (Gibbs and Kotsanas, 1986). Whereas the effect of experimental conditions on the basal metabolism of the in vitno perfused heart are now relatively well documented, studies of energy metabolism 0

1989

Academic

Press

Limited

212

G. Kotsanas

(biochemistry) under similar conditions are lacking. It was the aim of the present study to characterize the energetic state of rabbit hearts under conditions which, in our earlier study, were shown to result in differences in the basal Mpo, and CFR. In order to examine whether changes in the biochemical profile underly these differences in basal metabolism, the energy flux of arrested hearts was quantitated by measurement of oxygen consumption and the steady state concentrations of high energy phosphate compounds under different conditions. In addition, correlations were sought between the various parameters which are thought to describe the cellular energy state, e.g. the phosphorylation potential, and the basal M’C’o, . Methods

Rabbit hearts were perfused by the Langendorff technique as previously described (Gibbs et al., 1984; Gibbs and Kotsanas, 1986). Following cardiectomy, perfusion was initiated with a modified Krebs-Henseleit (KH) bicarbonate buffer of the following composition (IrIM) : NaCl 118.0, KC1 4.8, KH,PO, 1.2, CaCl, 2.5, MgSO, 1.2, and NaHCOs 24.8. Except where indicated all solutions contained 11 mM glucose as substrate, and 10 IU/l insulin. The pulmonary veins and vena cavae were ligated so that all coronary flow egressed via a flexible cannula that was inserted through the pulmonary artery and into the right ventricle. A small tube was also inserted into the left ventricle near the apex to drain any fluid accumulating there from either Thebesian flow or incompetence of the aortic valve. Experimental protocol All experiments were carried out at 27°C. Perfusion was initiated with KH at low coronary flow rates (perfusion pressures < 30 mmHg) while the heart was cannulated; it was then gradually increased to 75 mmHg and was maintained at this level throughout each experiment. During this period the hearts were allowed to beat spontaneously at their intrinsic heart rate (80 to 100 beat/min at 27°C) and any preparation that did not show regular coordinated contractions or gave indications of aortic valvular incompetence was rejected.

et al.

Perfusion was then initiated with one of the various cardioplegic solutions (see below) and was continued for 90 min once ventricular contractions had ceased. Measurements of the basal Mpo, and CFR were made at regular intervals (10 to 15 min) during this period, and upon completion of the final measurement a portion of tissue including both ventricles was compressed between stainless steel blocks precooled to the temperature of liquid N, . The unfrozen tissue that remained was blotted, weighed and dried to constant weight at 120°C to determine the tissue water content. Coronary arterial and venous samples were collected directly into 100 ~1 glass Hamilton syringes without exposure to air. The arterial perfusate was sampled from the aortic cannula 8 to 10 cm above the heart. The coronary venous effluent was collected by inserting the needle of the Hamilton syringe into the pulmonary artery cannula. Duplicate venous samples were routinely collected for analysis. To avoid oxygen exchange between the coronary venous effluent and the atmosphere only oxygen impermeable Tygon tubing was used for the pulmonary artery cannula and its length was kept as short as practical. In separate determinations the 0, content of perfusate emerging from a similar length of Tygon tubing attached directly to the aortic cannula was compared to the Oz content of the arterial perfusate. No loss of 0, could be detected with this length of Tygon tubing. Myocardial oxygen consumption (MOON) was calculated from the arteriovenous difference in 0, content and the coronary llow rate (CFR) and is expressed as m102/min/100 g wet wt of tissue. The values have been corrected for the increase in tissue water arising as a result of perfusion with saline solutions. Arterial and venous samples were analyzed for total 0, content in a Lex-0, Con TL oxygen analyzer (Lexington Instrument Corp., Waltham, Ma). To compensate for the low 0, content of physiological solutions and to enhance the accuracy of the measurements the sample volume was increased from 20 to 100 ~1 and the gain of the instrument was increased. A calibration factor was calculated prior to each experiment and all measurements were then corrected to allow for the

Energy

Metabolism

increased gain and sample volume. CFR was measured over 1 to 2 min by diverting the outflow from the pulmonary artery cannula into a graduated cylinder. Perfusion media Energy arrested

metabolism was studied in with the following solutions;

hearts

(1) KH with [KCl] elevated to 30 mM (high K KH) . (2) KH with [CaCl,] reduced to 0.10 mM and [KCl] elevated to 10.0 mM (low Ca KH). (3) Low Ca KH but with the Na salts of glutamate, fumarate and pyruvate (5 mM each) replacing glucose (low Ca KH + GFP). (4) Low Ca KH supplemented with 30 g/l of a high molecular weight dextran (Macrodex, MW 70 000, Pharmacia). This solution is designated low Ca KH + dextran. All solutions were maintained at 27°C and were equilibrated with a 95% Oz-5% CO2 gas mixture (PO, > 650 mmHg, final pH 7.3 to 7.4). Tissue processing Frozen tissues were extracted essentially as described by Williamson and Corkey (1969). The frozen tissue was finely powdered in a liquid N, cooled percussion mortar. It was then transferred to a precooled tube, quickly weighed and mixed with ice-cold 8% (v/v) HC104 in 40% ethanol (v/v). The resulting mixture was centrifuged to remove precipitated proteins and a predetermined volume of 3 M Kz CO, in 0.5 M triethanolamine was added to the supernatant to increase the pH to approximately 7.5. The insoluble KCIO, was then removed and the neutralized extracts were stored at -90°C until analyzed for the adenine nucleotides (ATP, ADP and creatine (Cr), creatine phosphate AMP), (CrP) and inorganic phosphate (Pi). Tissue extracts were analyzed for these compounds fluorometrically using the enzymatic methods of Lowry and Passonneau ( 1972). Calculation At the conclusion of wet weight to on a small sample tissue. This value

of cytosolic concentrations of each experiment the ratio dry weight was determined (50 to 70 mg) of the frozen was then used to determine

in the Arrested

Heart

213

the dry weight of the tissue that had been extracted. Total tissue metabolite levels deter,mined from enzymatic analyses are expressed as pmoles per gram wet weight of nonperfused heart using the dry weight of the extracted tissue and a wet weight to dry weight ratio of 4.05 for nonperfused tissue (determined in separate experiments, 12= 16). Cytosolic concentrations were calculated as described by Giesen and Kammermeier (1980). The cytosolic space of nonperfused rabbit heart was taken to be equal to 0.40 ml/g wet wt (Nayler et al., 1980). In the case of Pi allowance was also made for the amount contributed by perfusate adhering to the tissue at the instant of freezing. The cytosolic free [ADP] was determined using the equilibrium expression of the creatine kinase reaction

CATPI CCrl CADP’

= [CrP][H’]

x Xc,

where Kc, is the pH and [Mg2+]-dependent equilibrium constant of the creatine kinase reaction. 31P-NMR evidence (Nunnally and Hollis, 1979; Bitt1 and Ingwall, 1985) has confirmed that the reaction is at equilibrium in the perfused, arrested rat and rabbit heart. Intracellular pH at 27°C was assumed to be 7.2 lJacobus et al., 1982; Matthews et al., 1983). In the absence of accurate values for intracellular free [Mg’+] a value of Kc, =: 1.66 x ~O’/M at 1 mM free [Mg2+] was used (Lawson and Veech, 1979). The effects of temperature on Kc, were estimated from the Van’t Hoff equation using a value of -2.1 kJ/mol for the standard reaction enthalpy (Kuby and Noltmann, 1962) and the value of K cx at 27°C was then calculated to be 1.72 x log/M. The cytosolic free [AMP] was determined using the equilibrium expression of the myokinase reaction which also appears to be in equilibrium in vivo (Veech et al., 1979). It was calculated as

where the cytosolic free concentrations of ATP and ADP are used in the calculation. In the absence of specific information on the effects of temperature on the kinetics of the

214

G. Kotsanas

’ (b) F .G E g a, 5 5 .z ? : e 0"

lO.O8.0-

et al.

h ;

6.04.02.00.0

(6) (7) (7) (7) FIGURE 1. Basal oxygen consumption and coronary flow rate of arrested hearts. Values are mean & S.E. of the number of experiments shown in parentheses. *Significantly different (P < 0.05) compared to low Ca KH. n , High K KH; 0, low Ca KH; q low Ca KH + GFP; q , low Ca KH + dextran. reaction,

value [Mg’+] lations

and of the cytosolic free [Mg’+], a XmYO = 1.12 at 38°C and free = 1 mM was used in these calcu(Veech et al., 1979).

Results

of

Statistical analysis the All results are expressed as mean & standard error of the mean (S.E.). The unpaired Student’s t-test was used to test the effects of high K KH and low Ca KH solutions. Comparison between the three low Ca KH solutions was made using a one-way analysis of variance. If the F value indicated a significant difference then Tukey’s multiple range test was applied to determine differences. In all cases P values less than 0.05 were considered as indicating significance.

Basal Mvo, and CFR of arrested hearts The basal A&o, and CFR of arrested hearts perfused with different cardioplegic solutions are shown in Figure 1. There was a gradual decline in both n/rpo, and CFR during the first 30 min of arrest as has been previously reported for hearts perfused with saline solutions (Penparkgul and Scheuer, 1969; Gibbs and Kotsanas, 1986). Consequently only the mean values measured over the 60 to 90 min period of the arrest, when both parameters had plateaued to constant values, have been included. The basal A4p02 was two-fold higher, and CFR was three-fold higher for hearts arrested by Ca2+ depletion compared to K+ excess. Substitution of glucose with the

Energy

Metabolism

in

GFP substrate mixture in low Ca KH resulted in a moderate decline in CFR but stimulated the basal Mpo, by 25%. Addition ofdextran to low Ca KH led to a reduction of 35% and 75% for the basal Mvo, and CFR, respectively.

the

Arrested

215

Heart

Pi were all significantly lower with low Ca KH + GFP and low Ca KH + dextran (the reduction was 25% and 31% for ADP, 54% and 53% for AMP and 37% and 22% for Pi respectively). Cytosolic concentrations

Tissue metabolites in arrested hearts Arrested hearts were characterized by variability in the total tissue levels of ATP and CrP. The tissue ATP content which ranged between 2.3 to 3.3 pmol/g was within the range of reported values for perfused rabbit heart whereas tissue CrP levels tended to be higher (Gard et al., 1985; Malloy et al., 1986; Freeman et al., 1987). Differences in the tissue ATP and CrP levels across groups were not however accompanied by consistent changes in the levels of ADP and Cr and as a result the size of the adenine nucleotide and creatine pools depended on the perfusion conditions. It was also noteworthy that CrP accounted for a high fraction of the total creatine content. Under all conditions 92 to 94% of the total cellular creatine was in the phosphorylated form and this fraction was unaffected by the different perfusion conditions. Consequently the tissue CrP/ATP and CrP/Cr ratios were very high (see Table 3). While there was a tendency towards elevated tissue high energy phosphate levels in high K KH compared to low Ca KH arrested hearts (Table 1) the differences were not statistically significant. Compared to low Ca KH however, the tissue levels of ADP, AMP and TABLE

1.

Despite a two-fold difference in the basal Mvo, the cytosolic free concentrations of all metabolites were comparable between low Ca KH and high K KH arrested hearts (Table 2). The cytosolic [Pi] of hearts perfused with low Ca KH supplemented with GFP or dextran was 60% and 30% lower compared to low Ca KH alone (significant P < 0.05) but otherwise no significant differences in thle cytosolic concentrations of high energy phosphates were observed for hearts arrested by Ca*+ depletion. The estimated cytosolic free [ADP] ranged between 3.1 and 5.5 .ULMwhich is 250 to 3010 times less than the concentration that would be expected based on the total ADP extracteld from the tissue. The amount of ADP (in pmol) calculated to be free in the cytosol woulld therefore account for only 0.4% of the total, indicating that under these experimental conditions greater than 99% of the cellular ADP in the arrested heart is bound to intracellular proteins and/or compartmentalized. The cytosolic free [AMP] calculated from the free [ATP] and [ADP] and the equilibrium expression of the myokinase reaction was several orders of magnitude below thie concentration of other metabolites. Values

Tissue metabolites in arrested hearts Metabolite

Perfusion condition Low (n Low (n Low (n High (n

Ca KH = 7) Ca KH = 7) Ca KH = 7) K KH = 6)

(glucose) (GFP) (dextran) (glucose)

ATP

ADP

2.90 +0.23 2.27 f0.34 2.49 +0.22 3.30 + 0.42

0.59 +0.06 0.44a +0.02 0.41a f 0.03 0.74 +0.12

0.11 fO.O1 0.05” 20.01 0.05a kO.01 0.15 +0.04

of experiments (glucose).

indicated

Values are mean f S.E. of the number * P s 0.05 compared to low Ca KH ANP = adenine nucleotides.

AMP

level (p mol/g wet

wt)

CrP

Cr

Pi

Total ANP

10.78 &- 1.14 7.95 + 1.09 9.26 +0.78 13.61 +2.12

0.90 +0.19 0.52 f0.09 0.57 f0.08 1.08 +0.19

3.70 kO.38 2.32” +0.08 2.90” kO.26 4.27 +0.85

3.60 +0.26 2.76 +0.35 2.95 f0.24 4.19 +0.52

in parentheses.

Total creatine 11.68f 1.2S! 8.46 f 1.10 9.83 kO.81 14.69 +2.20

216

G. Kotsanas

TABLE

2. Cytosolic

concentration

of high

energy

et al. phosphates

Cytosolic

Perfusion

condition

Low (n Low (n Low

Ca KH = 7) Ca KH = 7) Ca KH (n = 7) High K KH

(glucose) (GFP) (dextran) (glucose)

(n = 6) Values

are mean +

a P I 0.05 compared ’ P I 0.05 compared

(mM)

CAMP1

x 10s

x 106

[Crpl

CCrl

6.33 + 0.56 4.80 + 0.84 5.34 +0.54 7.32 + 1.03

4.71 + 0.87 2.98 +0.75 3.12 +0.61 5.50

4.92 + 1.68 2.66 f 1.17 2.28 kO.69 5.32 f 1.88

26.30 f2.77 19.39 f2.66 22.58

2.19 +0.47 1.26 f0.21 1.39 f 0.20 2.64 +0.46

f1.09

of the number of experiments to low Ca KH (glucose). to low Ca KH (dextran).

S.E.

Energetic parameters in arrested hearts

parameters

concentration

CADPI

Table 3 lists the values of several parameters proposed to reflect the cellular energetic state. The [CrP]/[ATP], [CrP]/[Cr] and [CrP]/[Pi] ratios were unaffected by the method of arrest (Ca’+ depletion or Kf excess). In hearts arrested by Ca2+ depletion a change of substrate or addition of dextran also had no significant effect on these parameters. There was also no significant differ3. Energetic

free

hearts

CATPI

determined by this method were in the low nanomolar range and were unaffected by the different perfusion conditions.

TABLE

in arrested

in arrested

indicated

+ 1.89

33.18 +5.18

I31 5.13 + 0.86 2.04a*b f0.18 3.33” +0.57 6.41 f 1.91

in parentheses.

ence in the [ATP]/[ADP] ratio, the phosphorylation state of the adenine nucleotides or of creatine, or the AG of ATP hydrolysis between high K KH and low Ca KH arrested hearts. Significant differences in some parameters of energy metabolism related to the substrate and to the presence of dextran in the medium were, however, observed. The [ATP]/ [ADP][P,] ratio was three-fold higher with GFP compared to glucose and two-fold higher when dextran was added to low Ca KH. A similar trend was also observed for the [CrP]/[Cr][PJ ratio. A consequence of the higher adenine nucleotide phosphorylation

hearts Parameter

P-PI --CATPI Perfusion Low

Ca KH

(glucose) (GFP) (dextran)

(n = 7) High (n

K KH

= 6)

CATPI CADPI

15.66 + 3.33 17.68 + 3.06 18.05 +2.86 13.52 +2.29

1.71 f0.33

condition

(n = 7) Low Ca KH (n = 7) Low Ca KH

CC4 CCrl

(glucose)

4.14 f0.20 4.27 +0.30 4.33 f0.28 4.49 f0.15

x 10-3

1.93

+0.33 1.97 +0.31 1.47 +0.25

CATPI Wm&l 3.47 +0.52 9.348.b + 1.30 6.57= * 1.00 4.24 k1.46

CcrPl [ Cr] [Pi] x 1o-3/M

3.19 + 0.47 8.57a*b + 1.19 6.03” kO.91 3.89 + 1.34

AGATPC kJ/mol

-63.42 +0.41 -65.91avb f0.39 - 64.99’ kO.45 -63.09 +1.09

Values were calculated for individual experiments and the mean values ~s.E. have been tabulated. The number experiments is indicated in parentheses. ’ P I 0.05 compared to low Ca KH (glucose). s P I 0.05 compared to low Ca KH (dextran). ’ The free energy of ATP hydrolysis was calculated from the equation ACAT, = AG” + RT In [ADP][Pi]/[ATP] AG” was taken to be - 31.8 kJ/mol at pH 7.0 and lmn free [Mg”] (Lawson and Veech, 1979).

of

Energy

Metabolism

in the Arrested

217

Heart

1 (a)

12.0 s bx a = B 9 2

(b)

3.0-

+

k -

0.0

I

9.0 I

6.0-

I

-I I

$

I I

I

-.3 b x T & TY

Cc) 8.06.0-

+

4.0-

I

+ 2i Y

$

I

Z.O0.0

I 0.50 Oxygen

I I.50

I 1.00 consumption

(ml

FIGURE 2. Relationships between cytosolic free [ADP] (a), [ATP]/[ADP][Pi] (c) and the basal MVo2 in arrested hearts. Values are from Figure 1 and Tables KH; A, low Ca KH + GFP; A, low Ca KH + dextran.

state was that AGATp was 2.5 kJ/mol and 1.6 kJ/mol higher with low Ca KH + GFP and low Ca KH + dextran respectively. Correlations have been reported to exist between some of the energetic parameters shown in Table 3 and myocardial Mpo,. A linear and reciprocal relationship has, under certain conditions, been found between the logarithm of the [ATP]/[ADP][PJ and [CrP]/[Cr][P;] ratios and MPo,, while the cytosolic free [ADP] increases in parallel with Mpo, (Nishiki et al., 1978; Giesen and Kammermeier, 1980; Bunger and Soboll, 1986). It is clear from the results presented in Figure 1 and Tables 2 and 3 that there was no consistent correlation between the basal Mpo, which varied over a two-fold range and the cytosolic free concentration of any compound,

2‘.( IO

Oz/min/lOOgl

ratio (b), [CrP]/[Cr][Pi] ratio 2 and 3. 0, High K KH; 0, low Ca

or of any parameter that describes the tissue energy state. To illustrate this point Figure 2 shows the relationship between three selected parameters from Tables 2 and 3 and the basal Mro, . Discussion The aim of his study was to describe the energetic state of resting cardiac muscle under conditions where MVo2 and CFR varied over a significant range. The energy demand and the energy status of arrested hearts was estimated from measurements of Mvo, and from the steady state concentration of high energy phosphate compounds. It is evident from the results presented in Figure 1 that the basal metabolism of the perfused rabbit heart

218

G. Kotsanas

depends on the method of arrest (K+ excess or Ca*+ depletion), is sensitive to the nature of the oxidizable substrate, and to CFR, as was demonstrated when dextran was added to low Ca KH. Overall, the basal MVo, and CFR varied over a two to three fold range and their magnitude and the differences observed between the four different cardioplegic solutions are quantitatively similar to those previously reported from this laboratory (Gibbs and Kotsanas, 1986). Certain aspects of the energetic state of quiescent hearts as assessed by their high energy phoshate contents are noteworthy. Arrested hearts were characterized by variability in the total adenine nucleotide and creatine pools, and by high CrP/Cr ratios. The size of the creatine and adenine nucleotide pools was greater in high K KH perfused hearts than in hearts perfused with the different low Ca solutions, but did not correlate with either the basal Mpo2 or CFR. The reason for the variability is unclear. It is possible that some loss of high energy phosphates occurred as a result of trauma to the heart during isolation and preparation for perfusion (Vial et al., 1987). Another possibility is that perfusion with the low Ca solutions altered membrane permeability and resulted in loss of cellular constituents. A progressive loss of tissue high energy phosphates is observed when hearts are subjected to extracorporeal perfusion even with saline solutions of balanced ionic composition (Lee and Visscher, 1970; Vial et al., 1987). The concentration of CaCl, in the low Ca solutions was maintained at 100 pmol/l, well above the minimum level of 50 pmol/l below which damage to the cell membrane can occur (Grinwald and Nayler, 198 1). We have previously demonstrated that under similar conditions rabbit hearts are able to tolerate up to 3 h of perfusion with low Ca KH and then resume contractile activity upon return to normal KH containing 2.5 mmol/l CaCl, (Gibbs and Kotsanas, 1986). Nevertheless the possibility that changes in membrane permeability occurred cannot be discounted. The high CrP/Cr and CrP/ATP ratios arise as a consequence of (1) the high proportion of CrP relative to the total creatine, and (2) the low tissue ATP levels. Tissue ATP contents were similar to those that have been reported

et al. by others using chemical extraction techniques (Lee and Visscher, 1970; Freeman et al., 1987) but were lower than the in vivo ATP content of rabbit heart as estimated by tissue biopsy, or values derived from 31P-NMR studies (Gard et al., 1985; Malloy et al., 1986; Freeman et al., 1987). The high proportion of CrP to total creatine presumably reflects the relatively low metabolic demands of quiescent hearts. The concentrations of ADP and AMP calculated to be free in the cytosol were found to be well below the values expected if binding and compartmentation were ignored. Cytosolic free [ADP] ranged between 3.1 and 5.5 p~ while the AMP concentration was several orders of magnitude less with values in the range 2.3 to 5.3 nM. The high energy phosphate content of beating hearts was not determined ; however, the concentrations of free ADP and of AMP in perfused beating hearts are reported to be in the range 20 to 60 pM and 20 to 180 nM, respectively (Nishiki et al., 1978 ; Gard et al., 1985 ; Bunger and Soboll, 1986; From et al., 1986). Furthermore, the cytosolic free concentration of both nucleotides in perfused hearts appear to be directly related to metabolic rate (Bunger and Soboll, 1986). The lower values reported here are consistent with the observations of other groups who report lower cytosolic free concentrations of both nucleotides in arrested compared with beating hearts (Nishiki et al., 1978 ; Kauppinen et al., 1980). A direct result of the low cytosolic [ADP] is that the values of several parameters of energy metabolism (Table 3) were higher than is usually found in the perfused, beating heart (Kauppinnen et al., 1980; Matthews et al., 1983; Gard et al., 1985 ; Bunger and Soboll, 1986). The energetic state of arrested hearts was unaffected by the method of cardioplegia. There was no statistically significant difference in the tissue content of any metabolite or of its cytosolic free concentration in hearts arrested either by Ca*+ depletion or by K+ excess. The method of arrest also did not affect any of the energetic parameters examined, despite resulting in significant differences in the basal Mpo, and CFR. This observation is similar to that made by Penpargkul and Scheuer (1969) in rat heart but it differs from that of Nishiki et al. ( 1978) who found changes

Energy

Metabolism

in the tissue levels of some metabolites and a three-fold higher value of the phosphorylation potential in hearts arrested by Ca’ + depletion compared to hearts arrested by Kf excess. It may be argued that differences in the energetic state of the heart may go unnoticed for several reasons. Calculation of the cytosolic free [ADP] for example, requires that the intracellular pH and free [Mg”] under all conditions be accurately known, whereas the results shown in Tables 2 and 3 are based on the assumption that both remain unchanged. A recent report by Hoerter et al. (1986)) however, has shown that reduction of intracellular Ca 2+ of rat hearts p erfused with medium containing 80 pM CaCl, led to a progressive alkalinization of intracellular pH from 7.2 to 7.4. Under these circumstances the observed value of the equilibrium constant for the creatine kinase reaction would change. According to the equation of Lawson and Veech (1979) for the dependence of K,, with pH, the observed value would fall from 112 to 75. As a consequence the calculated value of cytosolic free [ADP] would increase by c. 50% and this would affect calculations of many cellular energetic parameters. Inspection of Table 3, however, shows that the [CrP]/[Cr][Pi] ratio which is practically independent of these considerations is similar for high K KH and low Ca KH and supports the view that energy metabolism does not appreciably differ between the two types of arrest. The substrate dependence of hearts arrested by Ca 2’ depletion was manifest as a higher basal Mpo2 and a significantly lower tissue level of Pi for hearts utilizing GFP. As a result of the lower cytosolic [Pi] the ratios of [ATP]/[ADP][Pi] and [CrP]/[Cr][Pi] as well as the AG,,, were all significantly higher than in hearts oxidizing glucose. Similar substrate dependent alterations of the myocardial energy state have been observed in the perfused, working heart preparation (Matthews et al., 1983; Starnes et al., 1985; Bunger and Soboll, 1986; Zweier and Jacobus, 1987). While there was no evidence in the present study of an increase in the tissue CrP content, as has been occasionally observed to occur in beating hearts when glucose is replaced with either pyruvate or noncarbohydrate substrates (Matthews et al., 1983; Zweier and

in

the

Arrested

Heart

219

1987), it appears that the energy Jacobus, status of the quiescent heart is altered in the same direction as that of the mechanically active heart by a change in substrate. A prediction of the near-equilibrium hypothesis of Wilson and colleagues is that ,a higher [ATP]/[ADP][Pi] ratio with GFP ought to be associated with an increase in the mitochondrial NADH/NAD+ ratilo (Erecinska and Wilson, 1978). The higher basal M?o, and adenine nucleotide phosphorylation potential with GFP is consistent with earlier work which has shown a higher resting heat rate and tissue NADH/NAD” ratio of rabbit papillary muscles utilizing pyruvate compared to glucose (Chapman and Gibbs, 1974). The improved energetic state of hearts perfused with the low Ca KH solution supplemented with dextran is an interesting observation. The three to four fold lower CFR and hence lower rate of 0, delivery under these conditions clearly had no detrimental effect, and in no way compromized the heart. The lower Mvo2 compared to low Ca KH is almost certainly related to the lower CFIR which lessens the distension of the myocardium (the “garden hose” hypothesis of Arnold et al., 1968). The reason for the improved myocardial energetic state with dextran is, however, unclear since dextran is an inert substance with no direct metabolic action. It may be related to the well documented property of colloid supplemented solutions maintaining membrane integrity and hence normal cellular structure (Armitage and Pegg, 197:7 ; Woo-Ming et al., 1980). In relation to this it was notable that the extent of tissue edema in hearts perfused with low Ca KH + dextrarn was lower than for all other solutions (wet weight to dry weight ratio for low Ca KH + dextran = 4.48 compared with a value of 5.17 for all other low Ca KH solutions). The relationships between cytosolic free [ADP], the phosphorylation state of the adenine nucleotides and of the creatine compounds and the Mflo2 of hearts operating over a wide range of workloads have been reported (Nishiki et al., 1978; Giesen and Kammermeier, 1980; Bunger and Soboll, 1986 ; From et al., 1986). An important observation to emerge from the present study is that in the in uitro arrested heart no single

220

G. Kotsanas

index of the myocardial energetic state consistently correlates with the basal Mpo, . This is clearly illustrated in Figure 2 where it can be seen that the phosphorylation potential can be elevated either by a change of substrate or by addition of dextran, manipulations which have quite opposite effects on the basal Mvo, and CFR. It was not the aim of this study to rigorously address the question of respiratory control and it is not implied that the results presented here argue against the general applicability of any of these parameters as reliable indices of myocardial respiratory rate in the working heart. In the past all have been found to correlate with respiratory rate either in isolated mitochondria or in perfused hearts. Differences in the basal MvoZ of arrested hearts in the present experiments encompassed a range sufficient to allow changes in energy metabolism to be detected. It may, however, not be surprising that these postulated indices of energy flux do not correlate with the basal Mpoz under all conditions, particularly as there is evidence in the literature that a substantial fraction of the oxygen uptake of the quiescent myocardium may in fact be unrelated to metabolic demand (ATP utilization). Challoner (1968) has provided data which suggests that 85 to 90% of the

et al. Mpo, of KG-arrested rat hearts is not coupled to oxidative phosphorylation, an observation later extended to rat hearts arrested by Ca ‘+ depletion by Penparkgul and Scheuer ( 1969). Furthermore, 20 to 30% of the oxygen uptake of the asystolic heart may be of non-mitochondrial origin. We are presently investigating whether the fractional contribution of mitochondrial nonphosphorylating respiration and nonmitochondrial respiration to the basal Mpo, is the same under the conditions prevailing in our experiments. Nevertheless under such circumstances Mvo, measurements may not be an accurate indicator of the real metabolic demand. The situation is made even more unclear by the recent calorimetric data of Buitenweg et al. (1987) which shows that the basal metabolic rate of very thin trabeculae (less than 0.6 mm diameter) taken from guinea-pig hearts can be as high as 34 mW/g wet wt (temperature = 37°C; 20 mM glucose and 2 mM pyruvate as substrates). This is equivalent to a basal Mvoz of approximately 10 m102/min/100 g, a value which is approximately half that of the working guinea-pig heart. If the biochemical status of such small preparations could be assessed a more searching test of the various energy indices could be made.

References ARMITAGE WJ, PEGC DE (1977) An evaluation of colloid solutions for normothermic perfusion of rabbit hearts: an improved perfusate containing Haemaccel. Cryobiology 14:428-434. ARNOLD G, KOSCHE F, MEISSNER E, NEITZERT A, LOCHNER W (1968) The importance of the perfusion pressure in the coronary arteries for the contractility and the oxygen consumption of the heart. Pfliigers Arch 321: 339-356. BITTL JA, INGWALL JS (1985) Reaction rates of creatine kinase and ATP synthesis in the isolated rat heart. J Biol Chem 260:3512-3517. BUITENWEG R, DAUT J, ELZINCA G (1987) Resting heat production at 37°C of superfused trabeculae isolated from guinea-pig heart. J Physio1386: 54P. BUNGER R, SOBOLL S (1986) Cytosolic adenylates and adenosine release in perfused working heart. Comparison of whole tissue with cytosolic non-aqueous fractionation analyses. Eur J Biochem 159: 203-213. CHALLONER DR (1968) Respiration in myocardium. Nature 217: 78-79. CHAPMAN JB, GIBBS CL (1974) The effect of metabolic substrate on mechanical activity and heat production in papillary muscle. Cardiovasc Res 8:656-667. CHAPMAN JB, GIBBS CL, LOISELLE DS (1982) Myothermic, polarographic, and fluorometric data from mammalian muscles. Correlations and an approach to a biochemical synthesis. Fed Proc 41: 176-184. ERECINSKA M, WILSON DF (1978) Homeostatic regulation ofcellular energy metabolism. TIBS 3:219-223. FREEMAN D, MAYR H, SCHMIDT P, ROBERTS JD, BINC RJ (1987) Advantages of perfluorochemical perfusion in the Biochim Biophys Acta 927: 35&358. isolated working rabbit heart preparation using 31P-NMR. FROM AHL, PETEIN MA, MICHURSKI SP, ZIMMER SD, UGURBIL K (1986) 31P-NMR studies of respiratory regulation in the intact myocardium. FEBS Lett 206: 257-261. GARD JK, KICHURA GM, ACKERMAN JJH, EISENBERG JD, BILLADELLO JJ, SOBEL BE, GROSS RW (1985) Quantitative 31P nuclear magnetic resonance analysis of metabolite concentrations in Langendorff perfused rabbit hearts. Biophys J #:803-813. GIBBS CL (1978) Cardiac energetics. Physiol Rev 58: 174-254.

Energy

Metabolism

in the Arrested

Heart

221

GIBBS CL, CHAPMAN JB (1979) Cardiac Energetics. In: Hmdbook ofPhysiolorThe Cardiovaxular System Section 2, Vol. 1, edited by RM Berne, N Sperelakis, SR Geiger. Bethesda, American Physiological Society, pp 775-804. GIBBS CL, WOLLEY G, KOTSANAS G, GIBSON WR (1984) Cardiac energetics in daunorubicin-induced cardiomyopathy. J Mol Cell Cardiol16,953-962. GIBBS CL, KOTSANAS G (1986) Factors regulating basal metabolism of the isolated perfused rabbit heart. Am J Physiol 250 (Heart Circ Physiol19) :H998-H1007. GIESEN J, KAMMERMEIER H (1980) Relationship of phosphorylation potential and oxygen consumption in isolate’d perfused rat hearts. J Mol Cell Cardiol12: 891-907. HOERTER JA, MICELI MV, RENLUND DG, JACOBUS WE, GERSTENBLITH G, LAKATTA EG (1986) A phosphorous-31 nuclear magnetic resonance study of the metabolic, contractile and ionic consequences of induced calcium alterations in the isovolumic rat heart. Circ Res 58: 539551. JACOBUS WE, PORES IH, LUCAS SK, KALLMAN CH, WEISFELDT ML, FLAHERTY JT (1982) The role of Intracellular pH in the Control of Normal and Ischemic Myocardial Contractility: A ‘iP Nuclear Magnetic Resonance and Mass Spectrometry Study. In: Intracellular pH: Its Measurement, Regulation, and Utilization in Cellular Functions. New York, Alan R. Liss, pp 5377565. KAUPPINEN RA, HILTUNEN JK, HASSINEN IE (1980) Subcellular distribution of phosphagens in isolated perfused rat heart. FEBS Lett 112: 273-276. KUBY SA, NOLTM~NN EA (1962) ATP-Creatine Transphosphorylase. In : The Enzymes, Vol. 6., edited by PD Boyer, II Lardy, K Mybrack. New York, Academic Press, pp 515-603. LAWSON JWR, VEECH RL (1979) Effects of pH and free Mg*+ on the K,, of the creatine kinase reaction and other phosphate hydrolyses and phosphate transfer reactions. J Biol Chem 254: 65286537. LEE YCP, VISSCHER MB (1970) Perfusate cations and contracture and Ca, Cr, PCr, and ATP in rabbit myocardium. Am J Physiol219: 1637-1641. LOCHNER W, ARNOLD G, MULLER-RUCHHOLTZ E (1968) Metabolism of the artificially arrested heart and of the gasperfused heart. Am J Cardiol22,299-3 11. LOISELLE DS, GIBBS CL (1979) Species differences in cardiac energetics. Am J Physiol237 :H9l%H98. LOISELLE DS (1987) Cardiac Basal and Activation Metabolism. In: Cardiac Energetics. Basic Mechanism and Clinical Implications, edited by R Jacob, Hj Just, Ch Holubarsch. New York, Springer-Verlag, pp 37-50. LOWRY OH, Passonneau JV (1972) A Flexible System ofEnqmatic Analysis. New York, Academic Press. MALLOY CR, MATTHEWS PM, SMITH MB, RADDA GK (1986) Influence of propranolol on acidosis and high-energy phosphates in ischaemic myocardium of the rabbit. Cardiovasc Res 20, 71s-720. MAITHEWS PM, BLAND JL, RADDA GK (1983) The temperature dependence of creatine kinase fluxes in the rat heart. Biochim Biophys Acta 763: 148146. NAYLER WG, SLADE A, VAUGHAN WILLIAMS EM, YEPEZ CE (1980) Effect of prolonged cc-adrenoreceptor blockade on heart weight and ultrastructure in young rabbits. Br J Pharmacol68: 363-371. NISHIKI K, ERECINSKA M, WILSON DF (1978) Energy relationships between cytosolic metabolism and mitochondrial respiration in rat heart. Am J Physiol234: C73-C81. NUNNALLY RL, HOLLIS DP (1979) Adenosine triphosphate compartmentation in living hearts: a phosphorus nuclear magnetic resonance saturation transfer study. Biochemistry 18: 3642-3646. PENPARGKUL, S, SCHEUER J (1969) Metabolic comparisons between hearts arrested by calcium deprivation or potassium excess. Am J Physiol217: 14051412. STARNES JW, WILSON DF, ERECINSKA M (1985) Substrate dependence of metabolic state and coronary flow in perfused rat heart. Am J Physio1249 (Heart Circ Physioll8) : H799H806. VIAL c, OWEN P, OPIE LH, POSEL D (1987) S’gt m ‘fi cance of release of adenosine triphosphate and adenosine induced by hypoxia or adrenaline in perfused rat heart. J Mol Cell Cardioll9: 187-197. VEECH RL, LAWSON JWR, CORNELL NW, KREBS HA (1979) Cytosolic phosphorylation potential. J Biol Chem

254: 653&6547. WILLIAMSON JR, CORKEY B (1969) Assays of Intermediates of the Citric Acid Cycle and Related Compounds by Fluorometric Enzyme Methods. In: Methods in Engmology Vol. 13, edited by SP Colowick, ND Kaplan. New York, Academic Press, pp 4395 13. WOO-MING M, Wu NM, LESTER JL, MALININ TI (1980) Water content of hearts perfused with crystalloid and nutrient solutions. J Mall Cell Cardiol12: 371-386. ZWEIER JL, JACOBUS WE (1987) Substrate-induced alterations of high energy phosphate metabolism and contractile function in the perfused heart. J Biol Chem 262:8015-8021.